Mamba
PulseAugur coverage of Mamba — every cluster mentioning Mamba across labs, papers, and developer communities, ranked by signal.
- instance of State Space Model 90%
- used by electroencephalography 90%
- used by magnetic resonance imaging 90%
- uses CNN 80%
- competes with attention 80%
- instance of long short-term memory 70%
- instance of State Space Models 70%
- developed electroencephalography 70%
- competes with CNN 70%
- competes with State space models: Univariate representation of a multivariate model, partial interpolation and periodic convergence 70%
- affiliated with magnetic resonance imaging 70%
- instance of gated recurrent unit 70%
18 day(s) with sentiment data
-
Nemotron-3-Super-120B-A12B achieves 504K token recall with Mamba+MoE architecture
NVIDIA's Nemotron-3-Super-120B-A12B model, a hybrid Mamba and Mixture-of-Experts architecture, has demonstrated perfect needle retrieval capabilities up to 504,000 tokens. This model utilizes Mamba layers to maintain a …
-
Prostate MRI segmentation gating behavior depends on backbone architecture
A new research paper explores the behavior of modality gating mechanisms in multi-modal segmentation for prostate cancer detection using MRI scans. The study, which involved extensive cross-validation across different b…
-
New Hybrid Mamba-Transformer Model Enhances EHR Representation
Researchers have developed HyMaTE, a novel hybrid model that combines Mamba (a State Space Model) and Transformer architectures to improve the representation of electronic health records (EHRs). This approach aims to ov…
-
Mamba models offer faster OCR but lag Transformer accuracy on historical texts
Researchers have benchmarked State-Space Models (SSMs), specifically Mamba, against Transformers and BiLSTMs for Optical Character Recognition (OCR) on historical newspapers. The studies indicate that while Mamba-based …
-
MambaADv2 framework enhances unsupervised anomaly detection with Mamba architecture
Researchers have introduced MambaADv2, a novel framework for unsupervised anomaly detection that leverages Mamba-based architectures. This approach aims to overcome the limitations of CNNs and Transformers by combining …
-
NVIDIA unveils efficient Nemotron 3 LLM family with hybrid architecture
NVIDIA has released two new large language models, Nemotron 3 Nano and Nemotron 3 Ultra, focusing on efficiency and advanced capabilities. Nemotron 3 Nano is a 30B-class model designed for private inference and agentic …
-
ITNet architecture unifies convolution, attention, and recurrence
Researchers have introduced ITNet, a novel neural network architecture that unifies convolution, attention, and recurrence into a single learnable integral transform. This architecture uses a learnable kernel, implement…
-
U$^2$Mamba network enhances salient object detection with nested U-structure
Researchers have introduced U$^2$Mamba, a novel U-structured network designed for salient object detection. This model leverages Mamba-based architectures to effectively model long sequences and incorporates multiscale …
-
SIMBA framework enhances weather prediction with bidirectional radiance modeling · 2 sources tracked
Researchers have developed SIMBA, a novel bidirectional framework for modeling hyperspectral infrared radiances from the FY-4A GIIRS instrument. This framework uniquely integrates atmospheric profile retrieval and radia…
-
New STORM framework enhances Mamba models by preserving spatial structure during token reduction
Researchers have developed STORM, a novel spatial-aware token reduction framework designed to address performance degradation in visual state space models like Mamba when subjected to token compression. Existing reducti…
-
Gaussian Mixture Attention offers linear-time sequence mixing
Researchers have introduced Gaussian Mixture Attention (GMA), a novel sequence mixing technique designed to overcome the quadratic scaling bottleneck of standard Transformer attention. GMA replaces explicit token-to-tok…
-
Deep learning models leverage energy features for improved surface classification in robotics
Researchers have explored the use of energy-derived features for surface classification in mobile robotics, comparing their effectiveness against inertial data. Utilizing deep learning models such as CNNs, RNNs, transfo…
-
Quantum-Enhanced Mamba-CNN Achieves 84.83% Accuracy in Crop Analysis
Researchers have developed a novel framework for crop field analysis using hyperspectral imagery, combining a multi-scale Convolutional Neural Network (CNN) with a Bi-directional Mamba module. This approach enhances spa…
-
Mamba and PPO achieve superior safety in spacecraft control
A new research paper explores the effectiveness of various recurrent neural network architectures and reinforcement learning algorithms for adaptive safety-critical control in spacecraft proximity operations. The study …
-
Reload-Mamba enhances semantic segmentation with novel state-space modeling
Researchers have developed Reload-Mamba, a novel framework designed to enhance multi-class semantic segmentation using Mamba-based state space models. This approach tackles the issue of response dilution in sequential p…
-
Shenzhen Big Data Institute's 4 AI research papers accepted by ICML 2026
The Shenzhen Institute for Big Data Research has had four of its research papers accepted by ICML 2026, a top-tier international conference in machine learning. Two of the papers introduce novel optimization techniques …
-
New AI Models Tackle Object Counting Challenges with Efficiency and Robustness
Researchers have introduced MambaCount, a new framework for text-guided open-vocabulary object counting that utilizes a Spatial Sparse State Space Duality (S^4D) block to overcome the limitations of Transformers in hand…
-
MambaH-Fit framework enhances point cloud normal estimation using state space models
Researchers have introduced MambaH-Fit, a new framework for point cloud normal estimation that utilizes state space models (SSMs). This approach aims to improve the modeling of fine-grained geometric structures, which c…
-
MamBOA architecture enhances video recognition with state-space models
Researchers have introduced MamBOA, a novel state-space architecture designed for video recognition tasks. This framework is backbone-agnostic, meaning it can integrate with existing CNN, Transformer, and Mamba architec…
-
New AI Methods Enhance Underwater Images and Object Detection
Researchers have developed new methods for enhancing underwater images, addressing issues like poor visibility, color distortion, and blur. One approach utilizes a deep unfolding network incorporating Mamba layers to ca…